Abstract not found

Geographical distribution is widely held to be a major determinant of evolutionary dynamics. Correspondingly, genetic programming theorists and practitioners have long developed, used, and studied systems in which populations are structured in quasi-geographical ways. Here we show that a remarkably simple version of this idea produces surprisingly dramatic improvements in problem-solving performance on a suite of test problems. The scheme is trivial to implement, in some cases involving little more than the addition of a modulus operation in the population access function, and yet it provides significant benefits on all of our test problems (ten symbolic regression problems and a quantum computing problem). We recommend the broader adoption of this form of “trivial geography” in genetic programming systems.

A number of learning models are commonly employed in the simulation of social be-haviour. These include population learning, lifetime learning and cultural learning. Pop-ulation learning allows populations as a whole to evolve over time, typically through a Darwinian model of natural selection. Lifetime learning allows individuals to acquire knowl-edge during their lifetimes and cultural learning allows individuals to pass this knowledge to their peers or the next generations. This work examines the effects of cultural learning on both the fitness and diversity of a population of neural network agents. A population employing population learning alone and one employing both population and cultural learning are assigned three benchmark tasks: the 5 bit-parity problem, the game of tic-tac-toe and the game of connect-four. Each agent contains a genome which encodes a neural network controller used by the agent to perceive and react to environmental stimuli. Results show that the addition of cultural learning promotes improved fitness and sig-nificantly increases both genotypic (the genetic make up of individuals) and phenotypic (the behaviour of individuals) diversity in the population.

This paper explores distance measures based on genetic operators for genetic programming using tree structures. The consistency between genetic operators and distance measures is a crucial point for analytical measures of problem di#culty, such as fitness distance correlation, and for measures of population diversity, such as entropy or variance. The contribution of this paper is the exploration of possible definitions and approximations of operator-based edit distance measures.

Self-assembly is a ubiquitous process in nature in which a disordered set of components autonomously assemble into a complex and more ordered structure. Components interact with each other without the presence of central control or external intervention. Self-assembly is a rapidly growing research topic and has been studied in various domains including nano-science and technology, robotics, micro-electro-mechanical systems, etc. Software self-assembly, on the other hand, has been lacking in research efforts. In this research, I introduced Automated Self-Assembly Programming Paradigm (ASAP 2), a software self-assembly system whereby a set of human made components are collected in a software repository and later integrated through self-assembly into a specific software architecture. The goal of this research is to push the understanding of software selfassembly and investigate if it can complement current automatic programming approaches such as Genetic Programming. The research begins by studying the behaviour of unguided software self-assembly, a process loosely inspired by ideal gases. The effect of the externally defined environmental

This dissertation concerns the principles and techniques for scalable evolutionary computation to achieve better solutions for larger problems with more computational resources. It suggests that many of the limitations of existent evolutionary algorithms, such as premature convergence, stagnation, loss of diversity, lack of reliability and efficiency, are derived from the fundamental convergent evolution model, the oversimplified “survival of the fittest” Darwinian evolution model. Within this model, the higher the fitness the population achieves, the more the search capability is lost. This is also the case for many other conventional search techniques. The main result of this dissertation is the introduction of a novel sustainable evolution model, the Hierarchical Fair Competition (HFC) model, and corresponding five sustainable evolutionary algorithms (EA) for evolutionary search. By maintaining individuals in hierarchically organized fitness levels and keeping evolution going at all fitness levels, HFC transforms the conventional convergent evolutionary computation model into a sustainable search framework by ensuring a continuous supply and incorporation of low-level building blocks and by culturing and maintaining building blocks of intermediate levels with its

microarray data that uses diverse ensemble

This paper examines the effects of lifetime learning on populations evolving genetically in a series of changing environments. The analysis of both fitness and diversity of the populations provides an insight into the improved performance provided by lifetime learning. The NK fitness landscape model is employed as the problem task, which has the advantage of being able to generate a variety of fitness landscapes of varying difficulty. Experiments observe the response of populations in an environment where problem difficulty increases and decreases with varying frequency. Results show that lifetime learning is capable of overall higher fitness levels and, in addition, that lifetime learning stimulates the diversity of the population. This increased diversity allows lifetime learning a greater level of recovery and stability than evolutionary learning alone.

Abstract. Cultural learning allows individuals to acquire knowledge from others through non-genetic means. The effect of cultural learning on the evolution of artificial organisms has been the focus of much research. This paper examines the effects of cultural learning on the fitness and diversity of a population and, in addition, the effect of selfadaptive cultural learning parameters on the evolutionary process. The NK fitness landscape model is employed as the problem task and experiments employing populations endowed with both evolutionary and cultural learning are compared to those employing evolutionary learning alone. Our experiments measure the fitness and diversity of both populations and also track the values of two self-adaptive cultural parameters. Results show that the addition of cultural learning has a beneficial effect on the population in terms of fitness and diversity maintenance. Furthermore, analysis of the self-adaptive parameter values shows the relative quality of the cultural process throughout the experiment and highlights the benefits of self-adaptation over fixed parameter values. 1

We hypothesize that the relationship between parameter settings, speci cally parameters controlling mutation, and performance is non-linear in genetic programs. Genetic programming environments have few means for a priori determination of appropriate parameters values. The hypothesized nonlinear behavior of genetic programming creates di culty in selecting parameter values for many problems. In this paper we study three structure altering mutation techniques using parametric analysis on a problem with scalable complexity. We nd through parameter analysis that two of the three mutation types tested exhibit nonlinear behavior. Higher mutation rates cause a larger degree of nonlinear behavior as measured by tness and computational e ort. Characterization of the mutation techniques using parametric analysis con rms the nonlinear behavior. In addition, we propose an extension to the existing parameter setting taxonomy to include commonly used structure altering mutation attributes. Finally we show that the proportion of mutations applied to internal nodes, instead of leaf nodes, has a signi cant e ect on performance.